Thanks. For us at least, there is no need to change the behavior, we just
wanted to understand the reasoning behind the Javadoc.
Den ons. 23. apr. 2025 kl. 20.36 skrev Matthias J. Sax :
> I guess that make sense. Would need to dig into the code (not super
> familiar with this part of the code bas
I guess that make sense. Would need to dig into the code (not super
familiar with this part of the code base), to fully confirm.
Not sure if it might not be possible to change though, if there is a
demand for it. -- Might also be a question about how complex such a
change would be.
My read i
Thanks, this was very useful.
> Well, the producer does not have x=1 in it's buffer any longer --
> otherwise, it our-of-sequence-error would not have be given to the
> application, and the producer could just retry internally.
Okay, so if I understand you, the reason the application has to repla
Thanks for providing more details.
It makes sense that you need to ensure that an "insert" comes before an
"update".
If I understand you correctly, that's what idempotence guarantees. If the
PUT is not appended to the log, it will be retried, but then thread 2 can't
have observed it.
Yes.
Thanks. What I'm looking for isn't that messages can't be interleaved, it's
something like causal order.
We're updating a database where every node has their own copy of the
database state (similar to RocksDB), by sending changes to that database
via Kafka.
In our application thread 1 is creating
The behavior I'm looking for is that if thread 1 sends message x=1, and
thread 2 sees message x=1 come through Kafka, thread 2 can't then write
message x=2 and have that be clobbered because thread 1's producer is
retrying the send since it didn't receive the ack in time.
This is not how it work
Hi,
If I understand correctly, the idempotent producer should be able to
guarantee that messages are not duplicated when the producer retries sends,
and as of https://issues.apache.org/jira/browse/KAFKA-5494, it should allow
for the producer to have multiple in flight requests at a time without
r
